It works for a while. If I export video only then I get to about 40 minutes. If I include audio then it crashes after only a couple of minutes.
How long are your renders that are working?
As far as being slower is concerned then yes but if you include the time taken to render the original DNXhR or equivalent before doing the DCP in DoM or easyDCP then I think the total time is about the same.
Thx for sharing, but my co-workers and I have all been making DCP on DCP-o-Matic for a while now... Anyway, it's always good to know...
*MacMini M1 16 Go - Sonoma - Ext nvme SSDs on TB3 - 14 To HD in 2 x 4 disks USB3 towers *Legacy MacPro 8core Xeons, 32 Go ram, 2 x gtx 980 ti, 3SSDs including RAID *Resolve Studio everywhere, Fusion Studio too *https://www.buymeacoffee.com/videorhin
Thom Britten-Austin wrote:I wonder what the guys over at EasyDCP will do now?
They probably aren't worried; the built-in DCP export won't do encryption, so EasyDCP isn't going to lose any business. It's just making DCP export easier for those of us who aren't able to spend the money for an EasyDCP license... anyone who'se using one of the free/open source DCP tools wasn't an EasyDCP customer yet anyway.
Rakesh Malik Cinematographer, VFX Artist, photographer, adventurer, martial artist http://WinterLightStudios.ca System: Asus ProArt 16/64GB/12 core Zen5/nVidia RTX 4070 8GB Nuke/Houdini/Resolve
Have now tried using a MC and it fails there too. Much later than in the PC but still falls apart after about 45 minutes into the program with 'failed to encode frame'
so something is wonky with your frame encoding process when encoding to jpg2000
Hope you can find it and fix it.
The test material was a 60 project in DNxHR 12 bit
beware of delivery DCP or Jpeg2000 or IMF: it pushes the CPU very strong consumption (here 100%). It must have a good FAN CPU and good ventilation otherwise: overheating
"Saying it is good, but doing it is better! " Win10-1809 | Resolve Studio V16.1 | Fusion Studio V16.1 | Decklink 4K Extreme 6G | RTX 2080Ti 431.86 NSD driver! |
Jean Claude wrote:beware of delivery DCP or Jpeg2000 or IMF: it pushes the CPU very strong consumption (here 100%). It must have a good FAN CPU and good ventilation otherwise: overheating
You can also limit the max frame speed in export
W10-19043.1645- Supermicro MB C9X299-PGF - RAM 128GBCPU i9-10980XE 16c 4.3GHz (Oc) Water cooled - 12x8TB SSD RAID5 internal (80TB) Decklink Studio 4K (12.4.1) Resolve 19.1.3 / fusion studio 19 GPU 3090ti drivers 512.59 studio
Thom Britten-Austin wrote:Jepp, tried that as well. Screwed it all the way down to 10, still falls to bits after about 65% i.e 40 minutes into the production
I'm pretty sure that with the setup at "max" you're getting not real-time but some half time (12-15) fps.
So, go down to 5, otherwise thete is no significant advantage.
Also you can render an intermediate dpx/dnx/ prores then do the dcp.
W10-19043.1645- Supermicro MB C9X299-PGF - RAM 128GBCPU i9-10980XE 16c 4.3GHz (Oc) Water cooled - 12x8TB SSD RAID5 internal (80TB) Decklink Studio 4K (12.4.1) Resolve 19.1.3 / fusion studio 19 GPU 3090ti drivers 512.59 studio
Has anyone SCREENED the DCPs using the Kakadu settings? The name's not exactly inspiring.
My very first Kakadu render was glitchy; same short film on a second pass looks clean.
GH ---------------------------------------------------- Greg Huson Secret Headquarters, Inc Post Production / Production Santa Monica, CA 323 677 2092 www.SecretHQ.com greg (at) SecretHQ.com
Glad Peter took my submission about building these tools into DR onboard, tho took a while. (2 years)
Anyway, I cannot seem to get 5.1 audio into a kakadu DCP render. Maybe I am not getting the output setup right. Has anyone been able to render 5.1 audio against DCP-kakadu?
I setup a 5.1 channel BUS, and select that as the audio channel to use in the render, then complains about EasyDCP needs even number of channels.. Maybe needs more work by the Devs. But if anyone has gotten it to work, drop me a line. I have not tried individual tracks, but that would take ages to setup and be a pain.
My test show Kakadu is 15-20% faster on my hi-core system. (24-cores) Kakadu is also the leader in J2k, been the most used and developed J2K codec. Most everyone else uses it or bases their version on the Kakadu code base. Its used mostly in Military and large cooperates doing a lot of image manipulation. But most of the DCP tools apart from EasyDCP(Franhofer based and developed), and DVS-Clipster, (Hardware implementation) use Kakadu for DCP encoding. Also very respected in medial (X-Rays and such). It is said to have the best bit to quality result. Tho, I understand the others may have cough up a lot.